The amniotic membrane forms a tough fluid-filled sac that protects the developing embryo and is essential for a successful pregnancy. Amniogenesis initiates early in human development as the embryo implants into the uterine wall: the inner cell mass first polarizes to form a pluripotent cyst with central lumen and subsequently, one half of this polarized cyst loses pluripotency markers and becomes squamous in nature (the prospective amniotic ecotderm) while the other side (the epiblast) remains pluripotent. Gastrulation begins on the epiblast side soon thereafter. These early post-implantation developmental steps are inaccessible to study in humans, leaving an enormous gap in our knowledge about amnion fate determination and formation of the amniotic sac, despite the central importance of these events to the survival of the developing embryo. A new in vitro model can help to close that gap: human pluripotent stem cells (hPSC), cultured in specific 3D conditions, form polarized pluripotent cysts that spontaneously self-organize into symmetric cysts composed entirely of amnion cells (90-95%) as well as asymmetric cysts that resemble amniotic sac-like structures (5-10%). Asymmetrically patterned cysts mirror Carnegie stage 5c human embryos and are called ?post-implantation amniotic sac embryoids? or ?PASE?. Cyst formation occurs progressively over five days in culture. Live imaging shows that asymmetric cysts arise from focal flattening at one pole of the cyst and laterally spreading of amnion fate; symmetric cysts arise when flattening occurs in a multi-focal pattern. Mechanistically, the initial trigger for amnion differentiation is mechanical and that this causes presumptive amnion cells to activate a BMP signaling program that is both necessary and sufficient for amniogenesis. At 5 days of culture, PASE contain distinct populations of amnion, epiblast and boundary cells; epiblast cells initiate EMT movements similar to gastrulation. We will exploit this robust in vitro model to accomplish the following goals: Aim 1) Explore how mechanical signals activate BMP signaling. Novel PiggyBac-based tools for genetic modification of hESC will aid in these studies. Aim 2) Establish the hierarchy of gene activation that results in amniogenesis and development of mature PASE. Single cell RNAseq will be used to dissect the transcriptional cascade that accompanies symmetry breaking, spreading amniogenesis, boundary formation and initiation of epiblast EMT movements. Aim 3) Functionally test transcription factors that control amnion fate. Genetic deletion and overexpression studies will be used to explore the role of several potential master regulators of amnion fate. Overall, the work proposed here will greatly accelerate the pace of discovery regarding critical but previously inaccessible post-implantation events and thus will have enormous implications for understanding early processes that impact embryonic development and human fertility.